This relates generally to imaging systems, and more particularly, to imaging systems that support pixel sub-sampling video modes.
Image sensors are commonly used in electronic devices such as cellular telephones, cameras, and computers to capture images. In a typical arrangement, an electronic device is provided with an array of image pixels arranged in pixel rows and pixel columns. Circuitry is commonly coupled to each pixel column for reading out image signals from the image pixels.
Image sensors for which all image pixels in the array are read out are referred to as operating in full resolution mode. To support high-resolution video processing, image sensors are often operated in a reduced resolution or sub-resolution mode in which the amount of data that is read out is only a fraction of the full resolution image size. The sub-resolution mode is implemented via a pixel array sub-sampling technique in which only fraction of the pixels are sampled for readout.
A key mode for high resolution video processing is the 3×3 sub-sampling mode in which the output image data is reduced by a factor of 3 in both the X (horizontal) dimension and the Y (vertical) dimension. A conventional way for implementing 3×3 sub-sampling involves accessing data from every third row (i.e., by ignoring data entirely from the two intervening rows). Skipping two-thirds of the rows in this way may be referred to as Y-Skip3.
In each row that is actually being accessed, data from each pair of adjacent pixels is read out and combined to produce intermediate pixel values. Adjacent pairs of intermediate pixel values may then be combined to produce final pixel values. Each group of six image pixels in a row generates two final pixel values using this approach. Generating two image pixel values for every group of six image pixels in this way may be referred to as X-Bin2 with 1.5 scaling. Implementing 3×3 sub-sampling via the combination of Y-Skip3 and X-Bin2/Scaling-1.5 may cause inaccuracies since two-thirds of the data is thrown out in the Y dimension and image aliasing may result from the X-Bin2 and scaling operation.
It may therefore be desirable to provide improved ways of implementing 3×3 sub-sampling.